diy solar

diy solar

Finally, the start of my 25kw Ground Mount grid-tie system

Pipe is stupidly expensive right now. When I installed my IronRidge system back in 2017, the 3" hot dipped galvanized pipe was $5.25 /foot.. Now I'm helping someone design a system and just got quoted $15.85 per foot.. WTF? That's ridiculous.

The shipping docs are all backed up so we're not getting much imported pipe and everyone is being forced to use the domestic stuff.. its better pipe than the crap that comes from China, but for a solar mounting system, it makes zero difference because you don't have to cut leak-proof threads.

I'm currently looking for another way to make the structure.. at this point, its actually a lot cheaper to just build a common 2x4 lumber frame, put a roof on it like a lean-to, and use roof mount solar hardware.
 
Take a look at the sunmodo pricing through CEDgreentech, it is much better priced.

my quoted price from CED was about 2/3 the cost of what is shown on their site
 
https%3A%2F%2Fprod.static9.net.au%2Ffs%2Fcd0a8684-9099-41c5-8b7a-5a03f9e01566


How about a split tilt array? Same principle as having split azimuth arrays, but with tilt as the variable.

I still think further consideration should be given to having split azimuth arrays despite the tariff regime you current have in place.
 
@wattmatters Are you suggesting I should consider going with split azimuth even though import/export is 1:1? If so, why and what would you propose?

edit: i'm not doubting or attacking, but I think it is great to hear opinions and become educated. I never turn that opportunity down.
 
Last edited:
If so, why?
I know you have 1:1 tariffs now but the question is will your tariff regime always be like that?

1:1 tariff regimes are becoming less common, especially in regions increasingly adding renewables to the grid supply. And once grid feed in is valued less than imports from the grid, then increasing the self consumption ratio of your solar PV becomes the best strategy which is aided by widening the duration of the production curve.

From what I recall you don't have a peak export power limitation, else that'd another reason to do it.
 
@wattmatters They don't really want me to produce more than 20% of what I use, but I think they pretty much base that off the yearly average since they don't care about my system and it certainly generates more power than I am using at any given time. Also, my neighbor's power is fed off the same transformer, so ultimately he would likely use any extra power I produce during the day before it actually hit the transformer and then went through the rest of their grid. That fact helps me with my system size as well since he kind of acts like a buffer.

@Brightside mentioned a SW and SE array.

This got me thinking a little about creating an equilateral triangle for the whole system.
2 Strings (32 panels) at 180*
2 Strings at 135*
2 Strings at 225*

I did the numbers on pv watts at 35* tilt on all (seems to be the best angle unless I went higher) and this system produced about 3000kw less over a year than having them all at 180*. This would be a bit more complicated to configure when building the physical arrays, but I'm sure it would provide me a much more even power from the time the sun came up until the sun went down throughout the day. It COULD possibly be about a breakeven since the the SE and SW may get more power generation on the backside in the morning/evening.

What type of array configuration would you propose?
 
Last edited:
One of the things I read was that the power generated from the backside will help warm the panels as well. This can help with the snow/ice situation. How much, I have not a clue. I wanted to go with the portrait setup, but the cost was a lot more on both the sunmodo and the unirac. The unirac system is way more expensive because you have to source your own pipe and pipe is VERY expensive right now. Sunmodo was an easy decision once seeing this.

I'm planning for 3' front edge as well. I think that is the highest configurable in the plan generators.

Obviously, take this with a grain of salt. Albedo numbers. It at least gives you an idea of different materials.
Item Values
Grass 0.15 – 0.26
Black earth 0.08 – 0.13
White sand, New Mexico 0.60
Snow 0.55 – 0.98
Asphalt pavement 0.09 – 0.18
Concrete pavement 0.20 – 0.40

If I could find this decently prices around here then I think this would maybe be some of the better rock or gravel you could use. If not, then I'll probably just go with 5/8" minus which is a grayish, probably providing albedo .1-.15 (similar to asphalt, but maybe a little better)
View attachment 69923
I just ran the Sunmodo calculator and only got a price difference of $70 between 2 up portrait and 4 up landscape on 64 panels. I got $14176 and $14103 respectively for the sunturf ground mount with 63" ground screws. I also saw in a video where they said to assume a 30% discount on the quote so this would put it closer to $10k. Did you run this on their calculator?

I like the idea of having everything you need delivered in a turnkey kit.

I just looked at some 2" and 3" schedule 40 galvanized steel piping and saw about $12/ft for 2" and $24/ft for 3" schedule 40 galvanized in 10' lengths.

What snowload and wind speed do you have to design for?

My panels will be in a field that will be dry grasses and dirt in the summer and maybe slightly greener with dirt in winter. So, I'll get something, and when snow is on the ground things should be pretty good.
 
@Brightside

You're right. Maybe it was Unirac that had the large difference. I input mine and it was $450 more. With what you're saying about the snow, it's probably worth going with the portrait to allow the snow to slide off easier.

My permitting papers say 85MPH and 39lb/sqf. ASCE7-10. Sunmodo comes out to be 110mph and 50lb/sqf.

edit: yes, it was the unirac ULA that got much much more expensive. I just used their tool again to check.
 
@wattmatters They don't really want me to produce more than 20% of what I use, but I think they pretty much base that off the yearly average since they don't care about my system and it certainly generates more power than I am using at any given time. Also, my neighbor's power is fed off the same transformer, so ultimately he would likely use any extra power I produce during the day before it actually hit the transformer and then went through the rest of their grid. That fact helps me with my system size as well since he kind of acts like a buffer.

@Brightside mentioned a SW and SE array.

This got me thinking a little about creating an equilateral triangle for the whole system.
2 Strings (32 panels) at 180*
2 Strings at 135*
2 Strings at 225*

I did the numbers on pv watts at 35* tilt on all (seems to be the best angle unless I went higher) and this system produced about 3000kw less over a year than having them all at 180*. This would be a bit more complicated to configure when building the physical arrays, but I'm sure it would provide me a much more even power from the time the sun came up until the sun went down throughout the day. It COULD possibly be about a breakeven since the the SE and SW may get more power generation on the backside in the morning/evening.

What type of array configuration would you propose?
That what I will have. I will have my original array of 8kW that I plan to add 8 of my new panels to(extend the existing racking out to support them), so I'll have about 12kW facing 180 degrees due south at 30 degree tilt.

The 2 new arrays will be in a completely different area that has a bigger horizon. These arrays will either be a V or a /\ angle. You could go with a /-\ or \ _ / (first one the center row would be all the at the top). I'm still trying to figure out whether I want \/ or /\ angles, as there are beneifits to both if you figure backside gain. But I think it's preferential to really try to maximize front side though.
 
@wattmatters

My consumption pattern throughout the 24 hour period is pretty stagnant as 2/3 of my usage is from computers/servers that run 24/7. I wouldn't say I will be WAY over powered, but I SHOULD have a little surplus at the end of the year.

My plan is that in about 10yrs when I hopefully can retire then I should be able to reduce consumption a bit on the computers and servers. At this point my panels will have degraded a bit as far as production goes. My power production will likely come with decrease power usage as I get older and reduce work that requires electricity. This should give my system an extremely useful lifespan (as far as 100% offsetting used electricity) as long as I don't have complete failures in the panels. I'm sure I will need to replace the inverters at some point.
 
@Brightside I suppose I never thought of inverting the angles of the SW and SE. I figured focus on the top side and if I could tweak a couple things to get a little more out of the underside then I would, but not going to make it a priority.

edit: I think if you put a little space between them the backside would be more beneficial.
 
I just looked at some 2" schedule 40 galvanized steel piping and saw about $12/ft for 2" 10' lengths.

This would be $9.50/ft for rigid


$6/ft for IMC. Thinner wall, maybe OK for uprights? (Use thicker wall for horizontals that get bending loads from rails.)

 
@Hedges

I put my system in ULA with portrait and it was $22k on unirac. Let's say $8 a foot for 800' pipe, that's $6400, plus $3250 in concrete. Even if the actual cost of the ULA was $15k, you're still way way more than sunmodo.

I think if it were last year then ULA would be the way to go. It's just so expensive right now that I can't justify it against the alternative. If sunmodo was close to ULA in price I may even cancel the whole project.

The CHEAPEST solution I could find would be the GFT (single post setup). That system on unirac's site is $13.7k. I'm sure it would be around $10k, plus $2-3k for concrete. Again, it'd be inline with sunmodo, but I think we all came to the consensus that the single post style mounts would ultimately be inferior overtime. Max angle is 30*

edit: thinking aloud. NOW if you could pay someone to pile drive the C channels then that would make it look very enticing.
 
Last edited:
This is a great article for calculating spacing between rows: https://www.cedgreentech.com/article/determining-module-inter-row-spacing

Assuming I calculated mine properly, it is showing 42' between arrays with a 35* angle and 37' at 30*

I am almost considering Uniracs GFT system with a 30* angle. It is cheaper and then I can likely hire a company to come out and pile drive the pilings, certify it, and then is all I have to do is assemble it. I could probably do that for about the same price as the sunmodo system. Unirac has a 20 or 25 yr warranty on the mounting system, so they must be confident in its ability.

ALSO, unirac provides WA certs as well, so that saves $300 too.

I think I'm going to really look into this.
 
Last edited:
I'm currently looking for another way to make the structure.. at this point, its actually a lot cheaper to just build a common 2x4 lumber frame, put a roof on it like a lean-to, and use roof mount solar hardware.
That is such a good idea, if I had thought of that I'd never have used a frame made out of pipe. Given the rocks in our soil if I had this to do over I would have poured a concrete slab and mounted the array on the slab. When it heaves in the winter who cares, it all stays flat. You could use white shingles on the roof and it would be good for bifacial panels. And the lumber would last forever under a shingled roof, it would almost never get wet. And if you boxed in the sides you would almost completely eliminate the uplift forces, you could design it more like the walls of a house instead of a sail.
 
Back when I started, ULA wasn't cheap but panels cost much more. Rebates were 50% (compare tax credits 28% today.)

I had purchased one complete ULA setup for a 3kW array.

Turned out they sent me standard rails rather than heavy-duty. Vendor and manufacturer made it right by applying the difference in price to additional standard rails. Because I'm not in such a high wind area it was OK. 18' rails on two 2" horizontal pipes. Each pair held eight, 5' x 2' 120W panels (or six 165W panels with length to spare.)

I purchased additional panel mount clips and Unirac hardware to mount horizontal pipes on top of vertical. Perhaps could have fabricated by welding two angles together.
I bought aluminum angle and U-bolts from McMaster Carr, cut and drilled to make mounts for rails on horizontal pipes.
With these, I made three more 3kW or so arrays.

More recently I purchased additional Unirac hardware when I've seen surplus. Slot bolt/nut sets and other pieces.
I bought some heavy duty rails. Only 17' long, would have liked 20'
My present arrays have PV panels in landscape mode, supported by 2 rails. What I want to do is put in portrait mode with one rail between each two columns of panels by the long edge. Panels can extend 18" or so past either end of the rails for 20' tall array (4 panels.) One problem I have to address is ensuring panel frames can't pull off the glass. Maybe strap short side of frames together and keep under tension.

Objective is to reduce number of rails required and fit number of panels desired. Seven rails on 21' horizontal pipes hold 6x4 array of 5' x 42" panels in 20' high 21' wide. Wired 8s3p.
This fits a whole number of PV strings in each array, which will have multiple orientations.
 
Back
Top